On Monday I read a fascinating piece by the Washington Post's Jeff Guo, "Death predicts whether people vote for Donald Trump," about how in previous primaries, "Donald Trump performed the best in places where middle-aged whites are dying the fastest." Trends in mortality rates among middle-aged whites are something I've been interested in thanks to the work of UIC's Jay Olshansky, so I looked at the CDC-reported death rates for whites ages 45-64 at the county level in Illinois. As I expected, they were higher downstate, and lowest in the relatively well-off collar counties. Here's what it looks like for 2014, showing the number of deaths per 100,000 population:
In the Republican vote-rich collar counties outside Cook County, Trump had 37 percent to 23 percent for Rubio, 21 percent for Kasich and 13 percent for Cruz. In Cook County, Trump had 30 percent support compared with 25 percent for Cruz, 21 percent for Kasich and 20 percent for Rubio. Throughout the suburbs, 5 percent were undecided.
The contest is closest in Illinois' Downstate 96 counties, which include some of the state's most rural and evangelical areas. There, Trump had 29 percent to Cruz's 27 percent, ahead of Rubio's 19 percent and Kasich's 15 percent. An additional 9 percent were undecided.
This isn't an outlier. The three polls done on the Illinois Republican presidential primary in 2016 all basically say the same thing—at least as far as Donald Trump goes.
First was the Paul Simon Institute, released on February 22. (I've removed Ben Carson, who did no better than seven percent, and "someone else," which peaked at one percent.) It's a small sample size, just 306 Republicans out of 1,000 registered voters in the entire poll, but the results are in keeping with the Trib's latest.
Jeb Bush dropped out on February 20, so the Capitol Fax/We Ask America poll, conducted on February 24, doesn't include him. It also doesn't include Ben Carson, who later suspended his campaign (I've removed "other," which does no better than six percent, in the collar counties). It's got a nice big sample size of 1,311 registered Republican voters.
Cruz, who wears his evangelical bona-fides proudly and who, despite being a senator, is on the outs with the GOP establishment, unsurprisingly fairs poorly in the Chicago suburbs/collar counties across all three polls. He does best downstate, likely because, as the Trib's Rick Pearson suggests, that's where the state's evangelicals tend to live.
The Barna Group has found that Cruz is favored among practicing Christians (church attendees who pray at least once a week) and evangelicals, but Trump is favored among "non-evangelical born-again Christians," "notional Christians," and Catholics.
In the collar counties—economically thriving, more Catholic, and relatively moderate in political terms—Trump is doing well. In the Trib poll, better than anywhere else; in the Capitol Fax poll, the first taken without Jeb Bush in the race, no worse than anywhere else.
It's unexpected. FiveThirtyEight's model predicts that "Trump will fare best in states and congressional districts with small shares of college graduates," while Rubio's "ideal path would rely on highly educated parts of the Mid-Atlantic, West Coast, and Great Lakes regions." Rubio does reasonably well across the polls in and around Chicago, but so does Kasich. In fact, the writers at FiveThirtyEight are keeping an eye on whether Kasich can pass Rubio in the race for the third-place, establishment-conservative podium. They have Rubio well ahead of Kasich in Illinois, with Kasich and Trump running competitively in Ohio; Kasich torched Rubio in Michigan last night. It's looking like the party can't settle on a safe mainstream choice; Mitt Romney, the failed standard-bearer of such in 2012, made robocalls for Kasich and Rubio in Michigan.
Meanwhile, as the campaign evolves, Trump's base seems to be broader than anyone expected. He's doing well across a lot of different demographics; as the Barna group reported, "among the adults aligned with the Republican Party, the Trump juggernaut seems to have won over a plurality of people from each segment—except evangelicals and practicing Christians."And as he does so, he's making it harder for the GOP to toss him in a brokered convention.
Perhaps the most interesting crosstab in any of the polls comes from the Capitol Fax poll, which looks at how often those polled voted for the Republican candidate in the last three presidential elections:
|1 of 3 Republican||45%||11%||18%||9%||12%|
|2 of 3||38%||15%||24%||9%||10%|
|3 of 3||33%||21%||22%||10%||10%|
Now that's an interesting pattern. Trump does better the less often voters have voted for the Republican candidate for President. Cruz's support runs in the opposite direction. Kasich and Rubio are kind of stuck in the middle.
It's a difficult profile for the GOP to turn down: a candidate who has broad support across what is arguably the most representative state in the country, and who appeals most to voters who have been least likely to vote Republican in the past three elections.
The Trib poll found that his unfavorable ratings are extremely high—and an anti-trump protest this Friday has expected attendance in the thousands—but that opposition is balanced by what appears to be broad appeal. If Trump can win Illinois, doing well across different regions of the state while pulling support from less-frequent Republican voters, he might make a deal the GOP can't refuse.